spaCy/spacy/lang/fi/tokenizer_exceptions.py

from ..tokenizer_exceptions import BASE_EXCEPTIONS
from ...symbols import ORTH, NORM
from ...util import update_exc


_exc = {}


# Source https://www.cs.tut.fi/~jkorpela/kielenopas/5.5.html
for exc_data in [
    {ORTH: "aik."},
    {ORTH: "alk."},
    {ORTH: "alv."},
    {ORTH: "ark."},
    {ORTH: "as."},
    {ORTH: "eaa."},
    {ORTH: "ed."},
    {ORTH: "esim."},
    {ORTH: "huom."},
    {ORTH: "jne."},
    {ORTH: "joht."},
    {ORTH: "k."},
    {ORTH: "ks."},
    {ORTH: "lk."},
    {ORTH: "lkm."},
    {ORTH: "lyh."},
    {ORTH: "läh."},
    {ORTH: "miel."},
    {ORTH: "milj."},
    {ORTH: "Mm."},
    {ORTH: "mm."},
    {ORTH: "myöh."},
    {ORTH: "n."},
    {ORTH: "nimim."},
    {ORTH: "n:o"},
    {ORTH: "N:o"},
    {ORTH: "nro"},
    {ORTH: "ns."},
    {ORTH: "nyk."},
    {ORTH: "oik."},
    {ORTH: "os."},
    {ORTH: "p."},
    {ORTH: "par."},
    {ORTH: "per."},
    {ORTH: "pj."},
    {ORTH: "puh.joht."},
    {ORTH: "prof."},
    {ORTH: "puh."},
    {ORTH: "pvm."},
    {ORTH: "rak."},
    {ORTH: "ry."},
    {ORTH: "s."},
    {ORTH: "siht."},
    {ORTH: "synt."},
    {ORTH: "t."},
    {ORTH: "tark."},
    {ORTH: "til."},
    {ORTH: "tms."},
    {ORTH: "toim."},
    {ORTH: "v."},
    {ORTH: "vas."},
    {ORTH: "vast."},
    {ORTH: "vrt."},
    {ORTH: "yht."},
    {ORTH: "yl."},
    {ORTH: "ym."},
    {ORTH: "yms."},
    {ORTH: "yo."},
    {ORTH: "yliopp."},
    {ORTH: "ao."},
    {ORTH: "em."},
    {ORTH: "ko."},
    {ORTH: "ml."},
    {ORTH: "po."},
    {ORTH: "so."},
    {ORTH: "ts."},
    {ORTH: "vm."},
    {ORTH: "srk."},
]:
    _exc[exc_data[ORTH]] = [exc_data]

# Source: https://kaino.kotus.fi/visk/sisallys.php?p=141
conj_contraction_bases = [
    ("ett", "että"),
    ("jott", "jotta"),
    ("kosk", "koska"),
    ("mutt", "mutta"),
    ("vaikk", "vaikka"),
    ("ehk", "ehkä"),
    ("miks", "miksi"),
    ("siks", "siksi"),
    ("joll", "jos"),
    ("ell", "jos"),
]
conj_contraction_negations = [
    ("en", "en"),
    ("et", "et"),
    ("ei", "ei"),
    ("emme", "emme"),
    ("ette", "ette"),
    ("eivat", "eivät"),
    ("eivät", "eivät"),
]
for (base_lower, base_norm) in conj_contraction_bases:
    for base in [base_lower, base_lower.title()]:
        for (suffix, suffix_norm) in conj_contraction_negations:
            _exc[base + suffix] = [
                {ORTH: base, NORM: base_norm},
                {ORTH: suffix, NORM: suffix_norm},
            ]

TOKENIZER_EXCEPTIONS = update_exc(BASE_EXCEPTIONS, _exc)